Qian L.

Key Skills

Software

AWS SageMaker

Top Subject Matter

Recommendation modeling

anomaly detection

experimentation/telemetry analytics

Top Data Types

Text

Top Task Types

Text Summarization

Object Detection

Question Answering

Segmentation

Classification

Text Generation

Freelancer Overview

Business Intelligence Engineer - Amazon Web Services. Brings 5+ years of professional experience across complex professional workflows, research, and quality-focused execution. Core strengths include Internal, Proprietary Tooling, and AWS SageMaker. Education includes Master of Science, Seattle University and Master of Science, Washington State University. AI-training focus includes data types such as Computer Code, Programming, and Text and labeling workflows including Evaluation and Rating.

Labeling Experience

Business Intelligence Engineer - Amazon Web Services

AWS SageMakerText

As a Business Intelligence Engineer at Amazon Web Services, you owned end-to-end telemetry and analytics data pipelines supporting multiple services and products. You developed data collection, storage, and processing workflows using core AWS technologies, ensuring reliable loading, transformation, and extraction. You applied statistical modeling and experimentation methods to drive product measurement and improvements using Python and SQL.• Built ingestion and transformation pipelines using S3, Redshift, Athena, and SageMaker to source, load, and extract data.• Developed an end-to-end LightGBM recommendation model with engineered features from 10+ SQL sources and weakly supervised labels, delivering strong precision and recall.• Designed A/B tests for a webpage experience and used results to improve onboarding and trial-to-paid activation outcomes.• Owned ETL processes and data requests across two services and eight products, maintaining data integrity and accuracy across big-data platforms.

2022 - Present

Business Intelligence Engineer at AWS (Feb 2022–Apr 2026): weakly supervised labeling and manual anomaly labeling for ML model evaluation and pipeline automation.

Owned and supported the end-to-end ML and analytics workflow that included creating and using weakly supervised labels and conducting performance review using manually labeled anomalies. Implemented recommendation-model training and validation steps that relied on label generation and evaluation outputs for business impact. Transformed raw signals into structured outputs used downstream by data pipelines and anomaly detection automation. • Developed a LightGBM recommendation model using weakly supervised labels from 10+ SQL sources. • Achieved 98% precision and 67% recall by evaluating label quality against outcomes. • Drove performance review with manual labeling of anomalies and computed matching matrices. • Created SQL post-processing scripts and automated the pipeline for repeated evaluation cycles.

2022 - 2026

Education

Work History